I'm trying run an a priori power analysis for an SEM and I need to know how many estimated parameters are included in the SEM. I'm having trouble understanding what exactly counts as an estimated parameter in an SEM. Based on my readings, I believe that the relationships between each examined variable count as estimated parameters. Is this correct?

My SEM will test the variable relationships depicted in the figure below, which leads me to believe that my SEM contains 6 estimated parameters (one for each relationship, shown via the arrows). However, I'm not feeling confident about my current knowledge. Any help is greatly appreciated. Thank you so much!

enter image description here


1 Answer 1


The following covariance structure related parameters would typically be contained in the model:

  • 4 structural path (regression) coefficients
  • 3 structural residual variances (1 for each dependent/endogenous variable in the structural model)
  • 1 factor loading (the other one is typically fixed for identification)
  • 2 residual (error) variances for the observed variables (Pride and Respect)
  • 1 exogeneous variable (Procedural Justice) variance

If the mean structure is included, there may be additional parameters (means for exogeneous variables and intercepts for endogenous variables).


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.